Age-period-cohort modelling of type 1 diabetes incidence rates among children included in the EURODIAB 25-year follow-up study.

AIMS: Specific patterns in incidence may reveal environmental explanations for type 1 diabetes incidence. We aimed to study type 1 diabetes incidence in European childhood populations to assess whether an increase could be attributed to either period or cohort effects. METHODS: Nineteen EURODIAB centres provided single year incidence data for ages 0-14 in the 25-year period 1989-2013. Case counts and person years were classified by age, period and cohort (APC) in 1-year classes. APC Poisson regression models of rates were fitted using restricted cubic splines for age, period and cohort per centre and sex. Joint models were fitted for all centres and sexes, to find a parsimonious model. RESULTS: A total of 57,487 cases were included. In ten and seven of the 19 centres the APC models showed evidence of nonlinear cohort effects or period effects, respectively, in one or both sexes and indications of sex-specific age effects. Models showed a positive linear increase ranging from approximately 0.6 to 6.6%/year. Centres with low incidence rates showed the highest overall increase. A final joint model showed incidence peak at age 11.6 and 12.6 for girls and boys, respectively, and the rate-ratio was according to sex below 1 in ages 5-12. CONCLUSION: There was reasonable evidence for similar age-specific type 1 diabetes incidence rates across the EURODIAB population and peaks at a younger age for girls than boys. Cohort effects showed nonlinearity but varied between centres and the model did not contribute convincingly to identification of environmental causes of the increase.

Core Standards of the EUBIROD Project. Defining a European Diabetes Data Dictionary for Clinical Audit and Healthcare Delivery.

BACKGROUND: A set of core diabetes indicators were identified in a clinical review of current evidence for the EUBIROD project. In order to allow accurate comparisons of diabetes indicators, a standardised currency for data storage and aggregation was required. We aimed to define a robust European data dictionary with appropriate clinical definitions that can be used to analyse diabetes outcomes and provide the foundation for data collection from existing electronic health records for diabetes. METHODS: Existing clinical datasets used by 15 partner institutions across Europe were collated and common data items analysed for consistency in terms of recording, data definition and units of measurement. Where necessary, data mappings and algorithms were specified in order to allow partners to meet the standard definitions. A series of descriptive elements were created to document metadata for each data item, including recording, consistency, completeness and quality. RESULTS: While datasets varied in terms of consistency, it was possible to create a common standard that could be used by all. The minimum dataset defined 53 data items that were classified according to their feasibility and validity. Mappings and standardised definitions were used to create an electronic directory for diabetes care, providing the foundation for the EUBIROD data analysis repository, also used to implement the diabetes registry and model of care for Cyprus. CONCLUSIONS: The development of data dictionaries and standards can be used to improve the quality and comparability of health information. A data dictionary has been developed to be compatible with other existing data sources for diabetes, within and beyond Europe.

Genetic risk for co-occurrence of type 1 diabetes and celiac disease is modified by HLA-C and killer immunoglobulin-like receptors.

The prevalence of celiac disease (CD) in patients with type 1 diabetes (T1D) has been reported to be 5-7 times higher than in the general population. Risk factors for co-occurrence of both diseases have not been entirely established. The aim of our study was to analyze possible impact of human leukocyte antigen (HLA) class I and killer cell immunoglobulin-like receptors (KIRs) on the co-occurrence of T1D and CD. We analyzed 67 patients with T1D, 68 patients with CD, 69 patients with both diseases (T1D+CD) and 130 controls. Statistical analysis was based on two tailed Fisher exact test with corrections for multiple testing. After stratification by DR3-DQ2, an association of HLA class I part of the COX haplotype (A1-B8-Cw7-DR3-DQ2) was not observed with each of the studied diseases separately, but it could be shown in case of the co-occurrence of T1D and CD. Only in the group of patients with coexisting diseases, the presence of HLA-C*07 (P = 8.65×10(-3) ) and HLA-B*08 (P = 0.03) but not HLA-A*01 increased the succeptibility. Our current data indicated that C*07, contributing C1 ligand (Pc = 3.67×10(-5) ) rather than B*08, that possesses no KIR ligand, could have an impact on the innate immunity rout of this susceptibility. The significant combination of C1-KIR2DL3 (Pc = 1.97×10(-4) ) observed in patients with coexisting diseases supports this hypotesis. Interestingly, no association was observed when C1 in combination with its stronger inhibitory receptor KIR2DL2 was investigated. Predominantly, weak inhibition in patients with coexisting T1D and CD could lead to a natural killer cell response, making them vulnerable for developing more than one autoimmune disease.

Paediatric screening for hypercholesterolaemia in Europe.

Different screening strategies are currently recommended to identify children with (familial) hypercholesterolaemia in order to initiate early lipid management. However, these strategies are characterised to date by low adherence by the medical community and limited compliance by parents and children. In a literature review, the authors assess which children should undergo screening and which children are in effect identified through the currently recommended strategies. Furthermore, the authors discuss the different screening tools and strategies currently used in Europe and what is known about the negative aspects of screening. The authors conclude that currently recommended selective screening strategies, which are mainly based on family history, lack precision and that a large percentage of affected children who are at increased risk of future coronary artery disease are not being identified. The authors propose universal screening of children between 1 and 9 years of age, a strategy likely to be most effective in terms of sensitivity and specificity for the identification of children with familial hypercholesterolaemia. However, this concept has yet to be proven in clinical practice.

An influence of HLA-A, B, DR, DQ, and MICA on the occurrence of Celiac disease in patients with type 1 diabetes.

Celiac disease (CD) is more common in individuals with insulin dependent diabetes mellitus (T1D) than in the general population. HLA class II molecules DQ8 (DQB1*0302-DQA1*0301) and DQ2 (DQB1*0201-DQA1*0501) have been identified as key genetic risk factors in both diseases. While DQ8 conveys a higher risk for T1D, DQ2 is more frequent in CD. Less is known about the contribution of HLA class I. The gut immune system has been implicated in the pathogenesis of both diseases. The MICA, which is mainly expressed in the gastrointestinal epithelium and recognized by gammadeltaT lymphocytes and natural killer (NK) cells via the NKG2D, might play a role. The aim of our study was to identify possible HLA class I and MICA alleles and conserved extended haplotypes as risk factors for the development of CD in T1D. Three groups consisting of 37 individuals with T1D and CD, 67 individuals with only T1D and 70 controls were analyzed. HLA class I and MICA alleles were determined using Luminex technology. An occurrence of CD in individuals with T1D was most significantly associated with B*08 (P = 7.3 x 10(-13)), contributing more than any of the HLA class II alleles (DRB1*0301, P = 5.00 x 10(-10); DQB1*0201, P = 7.65 x 10(-8)). Moreover, the association with CD became stronger when B*08(B*08-DQA*0501-DQB1*0201-DRB1*0301, P = 5.07 x 10(-12)) was present in the DRB1*0301-DQB1*0201-DQA1*0501 (P = 5.00 x 10(-10)) extended haplotype. We suggest a combined influence of alleles present in the MICA*008-B*08-A1-DR3-DQ2 extended haplotype on the development of CD in Slovenian individuals with T1D, where B*08 or/and a gene located close to it may play an important role, independently of HLA class II.

Motor activity during asymptomatic nocturnal hypoglycemia in adolescents with type 1 diabetes mellitus.

Nocturnal hypoglycemia is reported in 13%-56% of adolescents with type 1 diabetes mellitus. It may be asymptomatic in more than 50% of patients. No noninvasive method for detecting asymptomatic nocturnal hypoglycemia (ANH) has so far proven successful. The aim of the present study was to evaluate quantitative changes of motor activity by actigraphy during episodes of ANH in adolescents with type 1 diabetes mellitus. A total of 18 patients aged 10-16 years with a history of ANH were investigated. Blood was sampled at half-hourly intervals between 22.30 and 06.00 hours with a micropump, and an actigraph was fastened to the right wrist. Blood glucose concentrations were measured and compared to motor activity. Nocturnal hypoglycemia was recorded in 10 patients (55%), with blood glucose during periods of hypoglycemia of 3.00+0.17 mmol/l (range, 1.2-3.4 mmol/l), and duration of hypoglycemia of 1.95+1.34 hours (range, 0.5-5.0 hours). All periods of hypoglycemia were clinically asymptomatic. Regression analysis revealed a statistically significant linear correlation ( p=0.03) between blood glucose concentration and the respective 30-min activity counts. Activity counts in patients with nocturnal hypoglycemia were significantly (ANOVA, p<0.02) higher than in patients with normoglycemia. We conclude that low blood glucose was significantly correlated with an increase in motor activity as detected by actigraphy. This implies the possibility of noninvasive screening of asymptomatic nocturnal hypoglycemia.

The HLA-DRB, -DQB polymorphism and anti-insulin antibody response in Slovenian patients with type 1 diabetes.

A combination of specific HLA class II antigens and the presence of type 1 diabetes (T1D)-related antibodies has a high positive predictive value for T1D but low sensitivity. The aim of the present study was to determine the frequencies of HLA-DRB-DQB deduced haplotypes associated with susceptibility and protection in Slovenian patients with established T1D, to evaluate the relationship between the HLA-DRB1-QBP-DQB1 haplotypes and the presence of insulin autoantibodies (IAA) and glutamic acid decarboxylase antibodies (GADA), and to access the possible impact of polymorphic QBP promoters on this relationship. A cohort of 135 patients with T1D (age 17.5 +/- 7.0 years, duration of T1D 9.14 +/- 6.3 years) was investigated. HLA-DRB1 and DQB1 alleles were typed using the polymerase chain reaction (PCR)-reverse line blot method. QBP promoter region alleles were determined using PCR-sequence-specific oligonucleotide hybridization (SSO) and PCR-sequence-specific primers (SSP). IAA and GADA antibodies were determined by enzyme-linked immunosorbent assay (ELISA). The chi-square test with Yates' correction was used for statistical analysis. Deduced haplotypes DRB1*0301-DQB1*0201 (P = 0.0001, OR = 3.4), DRB1*0401-DQB1*0302 (P = 0.0001, OR = 29.8), and DRB1*0402-DQB1*0302 (P = 0.008, OR = 4.7) were significantly more common, and DRB1*1501-DQB1*0602 (P = 0.0001, OR = 0.03) significantly less common in the investigated cohort than in a Slovenian control group. The highest risk and the strongest protective HLA-DR-DQ haplotypes found in Slovenian patients with T1D did not differ from those found in other Caucasian populations. While the DRB1*0301-QBP2.1-DQB1*0201 haplotype, where QBP2.1 did not help to further distinguish DQB1*0201-possessing haplotypes in IAA-positive and IAA-negative patients, was strongly associated with the presence of IAA, the DRB1*0101-QBP5.12-DQB1*0501 haplotype, although not protective compared to the control population, was associated with an absence of IAA in the investigated cohort. It is suggested that there may be a combined influence of the QBP5.12 promoter and the DQB1*0501 functional molecule on reduced IAA production.

Incidence of childhood-onset Type I diabetes in Slovenia and the Tuzia region (Bosnia and Herzegovina) in the period 1990-1998.

AIMS/HYPOTHESIS: The incidence rate of childhood-onset Type I (insulin-dependent) diabetes mellitus in Slovenian children during the period 1990-1998 was studied and compared to that of children from the Tuzla region (Bosnia and Herzegovina). The secular trend for a 25-year period in Slovenia was also investigated. METHODS: The incidence data were obtained from the national Slovenian Type I diabetes register and from the local Type I diabetes register held in Tuzla. The ascertainment was based on the capture-recapture method and was estimated to be 100%. RESULTS: The age-standardized incidence of Type I diabetes mellitus for the age group 0-14 years for Slovenia was 8.54 per 100,000 (95 %-C.I. 7.5-9.5) person-years for both sexes. The incidence for boys was 8.03 per 100,000 (95 %-C.I. 6.7-9.4) and 9.12 per 100,000 (95% -C.I. 7.6-10.6) for girls. The age-standardized incidence in the Tuzla region was much lower: 3.03 per 100,000 (95 %-C.I. 2.0-4.1) for the whole group, 3.44 per 100 000 (95 %-C.I. 1.8-5.0) for boys and 3.21 per 100,000 (95 %-C.I. 1.6-4.7) for girls. A very low incidence of 0.8 per 100,000 in the youngest age group (0-4 years) was observed in the Tuzla region. The linear trend of the Type I diabetes incidence rate in Slovenia has been steadily increasing by 3.6% per year for the last 25 years. The incidence rate in the period 1974-1985 was significantly lower than that in the period 1986-1998 (p < 0.00026). CONCLUSIONS/INTERPRETATION: Although the incidence rate in Slovenia slightly increased during the period 1990-1998, the incidence rate in the Tuzla region remained at the pre-war level.

Seasonality of birth in children (0-14 years) with type 1 diabetes mellitus in Slovenia.

The aim of this study was to find out whether there is seasonality of month of birth of children with diabetes in Slovenia and if so whether it differs from that of the general population. A cohort of 849 children and adolescents (0-14 years) with type 1 diabetes mellitus born between 1956 and 1998 were included in the study. Monthly and seasonal patterns of birth of the patients with diabetes were compared with the pattern of normal live births (n = 1,345,921) and the pattern of disease onset. Statistical analysis was made using Student's t-test to compare the means between the four seasons of the year, and single cosinor analysis for a period of 12 months. The children and adolescents with diabetes had a statistically significant different seasonality of month of birth compared to that of the general population, and an opposite pattern from the seasonality of month of onset of disease. The observations made are in accordance with observations made recently in other countries and support the hypothesis that a virus infection transmitted by the mother to the fetus during the annual viral epidemic induces the autoimmune process in the pancreatic beta-cells in genetically susceptible individuals who will subsequently develop clinical diabetes during childhood.